Commit Graph
5036 Commits
Author SHA1 Message Date
DottaandPaperclip 672555ffbc Record current rich ACP qualification evidence and capability gaps
Separate source415 and source81 controller evidence from the unchanged e822 runtime. Retain failed attempts, scope the native denial pass, document pending plan settlement and exact command correlation, and distinguish implemented Pi notices from remaining rich-field gaps.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-30 00:48:33 -05:00
DottaandPaperclip 9d27311635 Refresh ACPX patch hashes in the frozen dependency lock
Regenerate pnpm 9.15.4 patch metadata without changing dependency versions or peer resolution. Bind both ACPX installations to their checked-in patches, including complete terminal diagnostics and rich protocol guards.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-30 00:44:30 -05:00
DottaandPaperclip 81c20ad227 fix: use non-secret text in the environment editor demo
Replace the static demo credential shape so bundled UI evidence does not trigger the secret scanner.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-30 00:07:46 -05:00
DottaandPaperclip 41503eb38f fix: avoid fresh attachment staging paths for empty wakes
Keep authorized residue scrubbing and unsafe-path checks while leaving untouched workspaces unchanged when no wake attachments were selected.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 23:16:23 -05:00
DottaandPaperclip 4597967f89 Handle durable cancellation before workspace failure recovery
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 23:15:13 -05:00
DottaandPaperclip 05d8b40a06 fix(evals): correlate denied Copilot tool target by origin
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 23:13:23 -05:00
DottaandPaperclip e822b614fc fix(runner): await cancellation proof after clean event EOF
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 22:25:41 -05:00
DottaandPaperclip 8b6136bf66 test(runner): align native selector fixture with Pi profile v9
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 21:40:17 -05:00
DottaandPaperclip 285073afe0 fix(runner): settle ACPX cancellation after verified retirement
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 21:40:17 -05:00
DottaandPaperclip 951b06d08a fix(runner): fence warm ACPX owners on runtime contract changes
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 21:40:17 -05:00
DottaandPaperclip d6ff17fd0d test(runner-e2e): isolate denied-target watchers from startup churn
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 21:40:17 -05:00
DottaandPaperclip 632e7c9802 docs(runner): record paid ACP qualification gaps
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 21:40:17 -05:00
DottaandPaperclip 4376dbe0be fix(evals): recognize native ACPX sidecar requests
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 21:40:17 -05:00
DottaandPaperclip e154e85394 test: scope Telegram lease fault to its fixture company
Fence the credential-reclamation hook used by the global recovery sweep so it cannot leave synthetic lease owners on earlier companies. Exercise a paused foreign receipt after the fault is armed and verify it settles without a leaked lease.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 20:12:20 -05:00
DottaandPaperclip 5605c350b2 fix(runner): retain native Cursor mode in shared session checkpoints
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 19:59:27 -05:00
DottaandPaperclip a1ccf64737 test: clarify runnerd bootstrap ID correlation
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 19:13:40 -05:00
DottaandPaperclip 794e90a258 test: require exact attested bootstrap reads in native qualification
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:54:12 -05:00
DottaandPaperclip b41ff0377b fix: attest single-file ACP bootstrap read targets
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:54:12 -05:00
DottaandPaperclip 1597325184 test(runner): bound remote receipts by the cell deadline
Reserve 15 seconds for public teardown, share remaining time across SDK/socket/host waits, subtract setup time from observer TTL, and bound close independently without extending qualification.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:27:54 -05:00
DottaandPaperclip ebdf4f6966 Integrate reviewed runtime prerequisite and retain follow-up validation rollup
Preserve the already-applied review repairs and remote extensions. Carry the current validation rollup in this qualification follow-up so the runtime prerequisite remains below the review file limit.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:27:00 -05:00
DottaandPaperclip fe92554788 Wire native Daytona qualification and warm ACP journeys
Withhold native tasks until exact owned observers are armed, preserve sealed retirement proof before environment teardown, and register explicit local and Daytona warm continuity cases.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:26:29 -05:00
DottaandPaperclip 84a52694af Preserve typed ACP read operations in native evidence
Keep setup reads distinguishable from unknown operations without retaining raw paths, arguments, or contents.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:26:29 -05:00
DottaandPaperclip e7abc92d91 Move validation rollup update into qualification follow-up
Keep the current validation rollup on the Daytona qualification branch while the prerequisite runtime PR stays below the review file limit. Runtime and tests are unchanged.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:23:26 -05:00
DottaandPaperclip 133793b419 test(runner): scope remote evidence around verified runtime state
Admit the actual staged runtime layout, retain sentinel and sibling mutation coverage, distinguish observed PRP identity from the sandbox lease, and bound setup readiness inside the caller deadline.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:22:19 -05:00
DottaandPaperclip 63eae696e7 test: bind Copilot remote protection to sealed sandbox evidence
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:20:13 -05:00
DottaandPaperclip baf70bf186 Refresh Pi backend test fixture to current candidate identity
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:18:14 -05:00
DottaandPaperclip 522ab2d002 Fix canonical Cursor plan delivery expectations
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:18:14 -05:00
DottaandPaperclip 04e9e9aacc test: count all Copilot origins and verify marker after cleanup
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:17:56 -05:00
DottaandPaperclip 11b2842d69 Refresh Pi backend test fixture to current candidate identity
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:16:53 -05:00
DottaandPaperclip 0ccdd4f03f Fix canonical Cursor plan delivery expectations
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:16:53 -05:00
Dotta 1d5196c8e8 Merge branch 'codex/runner-acp-native-protection' into codex/runner-acp-production
* codex/runner-acp-native-protection:
  test: count all Copilot origins and verify marker after cleanup
2026-09-29 17:16:53 -05:00
DottaandPaperclip 3f552d753a test: count all Copilot origins and verify marker after cleanup
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:16:51 -05:00
DottaandPaperclip 9da7a7be62 test(runner): bind remote native evidence to exact run leases
Arm bounded filesystem and process observers before publishing native actions, retain sealed file and retirement receipts before ephemeral lease teardown, and provide an independent attached-command fixture.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:15:25 -05:00
DottaandPaperclip d1bb5729f9 Add Pi remote native evidence and browser denial fixtures
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:13:41 -05:00
DottaandPaperclip fce6e109ad Prepare Cursor native Daytona flows with owned remote evidence
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 17:13:41 -05:00
DottaandPaperclip 921cae91ad Keep Cursor mode defaults independent of qualification narrowing
Compute the selected profile mode before the literal qualification guard narrows pending providers out of the selectable union.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:54:06 -05:00
DottaandPaperclip 7fc2da7d34 Preserve Cursor command and denial evidence in both ACP transports
Observe only the active native prompt and bind permission delivery to the provider response write. Keep evidence failures separate from permission and terminal authority.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:49:15 -05:00
DottaandPaperclip fb788a9145 Keep non-command Cursor permissions from invalidating shell evidence
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:48:13 -05:00
DottaandPaperclip 95faa1ee8b Prove Cursor native denial command and cancellation boundaries
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:45:15 -05:00
DottaandPaperclip a9e6757255 docs(runner): separate revised candidates from historical paid evidence
Record Cursor5, Copilot5 and Pi8 deterministic validation, retain all earlier failures and identify fresh qualification gates.

Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-09-29 16:41:43 -05:00
DottaandPaperclip c22526c266 test(runner): register native Cursor interactions and cleanup proof gates
Add four explicit candidate cells with mode-specific native callback evidence. Run independent cleanup assertions before grading and preserve failures while closing every observer.

Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-09-29 16:41:43 -05:00
DottaandPaperclip 96ca6ce1e7 Clarify native fixture evidence requirements
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:41:43 -05:00
DottaandPaperclip 879eecdbb1 Prepare native Cursor callback and durable denial Product flows
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:41:43 -05:00
DottaandPaperclip 633ce41aa7 fix(runner): admit the exact Pi profile 8 native identity
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:30:28 -05:00
DottaandPaperclip a97b626081 fix(runner): preserve Pi native assistant boundaries and notices
Pin Pi profile 8 to native message provenance, preserve explicit empty starts and ends, and keep replay/history and native notices out of final assistant aggregation. Validate fresh, warm and loaded SDK sessions and exact native closure bytes.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:30:28 -05:00
DottaandPaperclip ceb28bdba4 fix(server): bind Cursor mode at recovery identity boundaries
Reject missing, foreign, or changed observed modes before accepting suspended checkpoints or interaction-driven identity transitions.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:27:18 -05:00
DottaandPaperclip c867cff960 fix(runner): bind candidate pack to Cursor v5 mode admission
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-09-29 16:25:03 -05:00
DottaandPaperclip f5546efbaa Reject Cursor mode drift in native configuration updates
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:25:03 -05:00
DottaandPaperclip 5013666821 fix(runner): bind Cursor mode across Rust session recovery
Require explicit Cursor modes, preserve them through open and checkpoints, and reject mode drift. Bind the current Cursor v5 identity while rejecting historical v4.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:24:09 -05:00
DottaandPaperclip 4af314a055 Admit and bind Cursor native session modes in profile v5
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-09-29 16:23:41 -05:00